Share
3,322 Posts.
lightbulb Created with Sketch. 103
clock Created with Sketch.
20/11/17
15:54
Share
Originally posted by grasshopper05
↑
I considered this possibility but if they changed the delta so that the counter was increasing by 2100 per day instead of 2000 then the way it is currently calculated would mean there would have been a huge jump as it would then apply the 100 increase from the start date which is ~450 days ago so it would result in a 45k jump. Also doesn't seem logical you would update the rate of increase and not update the starting user number and starting date.
In terms of why they did it this way is not so much being easier, making a query to the database to get user numbers would probably be quicker to code then needing to do all the calculations however it comes to system resources. Making a query to the database to update the user numbers every time someone visits the site is a waste of resources. Sending some static numbers to the visitors and then having the visitors computers do all the computing to get some numbers minimises the processing required on Aconex's end.
I will add that it does appear that the way it was built was so that it would be easy to update and maintain the tickers. The code snippet I attached above in on the main page whereas the calculations and initialisation are contained in a separate javascript file. Perhaps the person that originally built the tickers is no longer at Aconex as it would easy to update the starting numbers, date and increment amounts.
As for the accuracy the issue is that it is a static increase e.g. 2000 per day. Given it is approaching 18 months later and we know Aconex is growing 15-20% per year it stands to reason that the growth in users would also increase at roughly similar rates. When it was first built they put in a 16.6% growth rate or 2050 per day. If they have continued to grow at this rate then their user numbers would be 200k higher than what the ticker states now with the gap growing exponentially over time.
Anyway thanks to Just_a_guy for pointing it out as if they were live numbers being pulled it would have been a great source of information to monitor and track to see potential trends in usage.
Expand
Maybe you should offer your coding services to Aconex!
Appreciate you pointing this out. I thought it was a bit bizarre that the rate was very consistent across users mail and documents, but didn’t have the foresight to think it was simply coding.
I certainly wasn’t using it as a proxy for buying or selling my shares but I like tracking google trends etc none the less. Incidentally goggle trends seems to be trending up, which I hope is not simply coding!